KR20200068052A - Method for diagnosis and control of diseases and insect pests using multiple camera module - Google Patents

Abstract

The present invention provides a method for diagnosis and control of diseases and insect pests using a camera module which automatically diagnoses diseases and insect pests and automatically transmits a variety of information for controlling diseases and insect pests to a user. The method comprises (a) a step in which images acquired by a plurality of camera modules are transmitted to a diagnosis server (500) by an image collection unit (400); (b) a step in which the diagnosis server (500) compares the images transmitted in the step (a) with a plurality of disease and insect pest images stored in a disease and insect pest database (550) to check a disease and insect pest image with the highest accuracy and check disease and insect pest information mapped to the disease and insect pest image; (c) a step in which the diagnosis server (500) transmits the checked disease and insect pest information to a user terminal (800); (d) a step in which the diagnosis server (500) transmits the images transmitted in the step (a) and the disease and insect pest information checked in the step (b) to an expert terminal (900); and (e) a step in which a deep learning server (700), if disease and insect pest names different from disease and insect pest names included in the disease and insect pest information are inputted from the expert terminal (900), maps the different disease and insect pest names to the images transmitted in the step (a) to store the different disease and insect pest names in the disease and insect pest database (550) and store the different disease and insect pest names with a probability to which a preset expert weight (x) is applied. If the step (e) is repeated for the same disease and insect pest image by different expert terminals (900), the probability is accumulated to increase.

Description

Method for diagnosis and control of diseases and insect pests using multiple camera module}

The present invention relates to a method of diagnosing and controlling pests by combining IT technology with agricultural technology, and more specifically, installing a plurality of camera modules in a facility such as a vinyl house, and automatically diagnosing pests through the acquired images. This is a method of increasing accuracy by deep learning by repeating this process.

Timeliness is important for identifying pests of crops and diagnosing and controlling them. If the time is delayed, pests may become widespread and difficult or impossible to control, more pesticides may be used, or the yield may be greatly reduced.

In order to solve this, a variety of techniques have been developed to identify and diagnose pests in real time and to provide control technologies together.

Korean Registered Patent No. 10-1156594 provides a technology for real-time diagnosis of pests by establishing a separate pest database and comparing crop images with real-time crop images. Korean Registered Patent No. 10-1118245 provides an apparatus for analyzing pest distribution by installing a sticky plate at a fruit cultivation site and photographing it in real time.

The above patents have the following problems.

First, it provides information on only some of the various pests that occur when growing crops. In the former prior art, since only the crop is photographed, it is difficult to identify a pest with a small time remaining in the crop. The latter conventional technique only captures trapped worms, making it difficult to identify the disease that has occurred on the crop. In addition, it is impossible to identify insects that are not attracted to the trap. However, it is also unrealistic to install a large number of cameras in large-scale cultivation facilities such as facilities.

Second, it is difficult to guarantee pest accuracy. When diagnosing a pest through an image comparison method, there is a possibility of an incorrect diagnosis. Since the pest is not used, the pest is not controlled and only the crop yield may be lowered.

Third, if a new pest is discovered or a new control method is developed, it is difficult to immediately manage the data.

KR 10-1156594 B KR 10-1118245 B

The present invention has been devised to solve the above problems.

Specifically, it is intended to provide a system in which a pest is automatically diagnosed and various information for controlling it is automatically transmitted to a user.

In addition, to increase the accuracy, we apply a deep learning technology, automatically recommend the installation location change to secure the optimal image by the camera, and propose a method for identifying pests specific to crops. In particular, I would like to propose a method that can be machine-learned because it can actively accept user's opinion and expert opinion.

In addition, it is intended to propose a method capable of active diagnosis and control by flexibly introducing new pests or developing new control methods.

An embodiment of the present invention for solving the above problems, (a) the image acquired from a plurality of camera modules is transmitted to the diagnostic server 500 through the image collection unit 400; (b) the diagnostic server 500 compares the image transmitted in the step (a) with a plurality of pest images stored in the pest database 550 to identify the pest image with the highest accuracy, and the pest mapped therein Checking information; -The pest information includes the pest name (c) the diagnosis server 500 transmitting the identified pest information to the user terminal 800; (d) the diagnosis server 500 transmitting the image transmitted in the step (a) and the pest information identified in the step (b) to the expert terminal 900; And (e) when a pest name other than the pest name included in the pest information is input from the expert terminal 900, the deep learning server 700 maps the other pest name to the image transmitted in step (a). And storing it in the pest database 550 with the probability of applying the expert weight (x) set, and repeating the step (e) in different expert terminals 900 for the same pest image. In the case, the probability is cumulatively increased, providing a method for diagnosing and controlling pests using a camera module.

In addition, (c1) when the user terminal 800 inputs a pest name other than the pest name included in the pest information received, the deep learning server 700 is different from the image transmitted in the step (a) Mapping the pest name and storing it in the pest database 550, but further comprising the step of storing with a probability of applying a set user weight (y), the different user terminal 800 for the same pest image (c1) If step) is repeated, the probability is cumulatively increased, and it is preferable that the preset user weight (y) is lower than the preset expert weight (x).

In addition, in step (b), (b1) the diagnostic server 500 compares the images transmitted in step (a) with a plurality of crop images stored in the pest database 550, whichever has the highest accuracy. Checking the crop image of, and checking the crop name mapped to this; (b2) The diagnostic server 500 selects a plurality of pest images mapped with the pest name stored in advance as corresponding to the crop name identified in step (b1), and transmits the selected pest images and the step (a). Checking the pest image with the highest accuracy by comparing the images; And (b3) it is preferable that the diagnostic server 500 includes the step of checking the pest information mapped to the pest image identified in step (b2).

In addition, in the step (b), when the diagnostic server 500 confirms that there is no pest image with the highest accuracy and the image transmitted in the step (a), the step (c) includes the diagnosis server 500 It is preferable that the) further comprises the step of transmitting the pest-free information to the user terminal 800.

In addition, steps (a) to (e) are repeated every predetermined period, and it is preferable that the preset period can be changed through the user terminal 800.

In addition, the installation coordinates of each of the plurality of camera modules are stored in the camera control server 600, and after the steps (a) to (e) are repeated, the camera control server 600 may control each camera module. It is preferable to further include the step of checking the frequency at which the pest image is identified and transmitting the information recommending moving the plurality of camera modules toward the installation coordinates of the camera module having high frequency to the user terminal 800. .

In addition, it is preferable that the plurality of camera modules include a trap camera module 100, a prediction camera module 200, and a stray light insect repellent camera module 300.

In addition, the trap camera module 100, a hollow trap body 110, a trap plate 120 provided on an upper side of the lower surface of the trap body 110, a trap camera for directing the trap plate 120 130, and a transmitting and receiving unit 140 for transmitting the image captured by the trap camera 130 to the image collection unit 400, the image obtained in step (a) is the trap camera 130 It is preferable to include an image captured by ).

In addition, it is preferable that the pest information includes the pest name, the drug product name, the pesticide product name including the drug product name, and the pesticide use method of the pesticide product name.

In addition, (f1) the expert terminal 900 further inputs a pest image to the pest database 550; And (g1) it is preferable that the expert terminal 900 further comprises the step of mapping and storing the pest information pre-stored in the pest database 550 to the further pest image.

In addition, (f2) the expert terminal 900 selects any one pest image stored in the pest database 550; And g2) the expert terminal 900 further comprises mapping new pest information to the selected pest image and storing the pest information in the pest database 550.

In addition, (f3) the expert terminal 900 further inputs a pest image to the pest database 550; And (g3) the expert terminal 900 further comprising the step of mapping new pest information to the selected pest image and storing it in the pest database 550.

In addition, when a pest image or pest information is further input into the pest database 550, it is preferable that the inputted content is transmitted to the user terminal 800 and other expert terminals 900.

In addition, in step (b), (b4) the diagnostic server 500 compares the images transmitted in step (a) with a plurality of pest images stored in the pest database 550, and has two or more of the highest accuracy. It is preferable to further include the step of identifying the pest image and confirming two or more pest information mapped to each of them.

In addition, the step (c) may include: (c2) the diagnosis server 500 transmitting two or more pest information identified in the step (b4) to the user terminal 800; And (c3) it is preferable that the diagnostic server 500 further comprises the step of identifying the common drug item name in the two or more pest information identified in step (b4), and further transmitting it to the user terminal 800. Do.

According to the method according to the present invention, only by installing the camera module and periodically retrieving the image, pests of the crop can be confirmed in near real time, and the control method can be provided in real time to the farmer who is the user. Accordingly, when a significant time has elapsed since the occurrence of the pest, it is possible to fundamentally prevent the case of using the inaccurate control method, etc. It is possible.

To increase diagnostic accuracy, pests are identified in a variety of ways. For example, by acquiring images by attracting insects in different ways, such as pheromones and manned light, we want to check the images of all insects in the facility, and we also install a camera that predicts the crop itself to separate crops that are not caused by insects Bottles are also identifiable.

When performing the method according to the present invention, diagnosis and control with high accuracy are possible. In particular, if the diagnosis server makes an incorrect diagnosis by actively reflecting the opinions of experts and users, it is possible to correct this, as well as reduce the possibility of incorrect diagnosis in similar situations in the future by performing machine learning through deep learning technology.

When a new pest is discovered or a foreign pest is introduced, as well as a new control method is developed, new information can be input and applied in a simple manner by the method according to the present invention, thereby enabling active and flexible response. Pest information and control information newly input by experts may be actively announced to the user.

When installing in a farmhouse, it may be impossible to install many camera modules for economic reasons. In this case, if the installation positions of the few camera modules are incorrect, the pest may not be quickly diagnosed. The method according to the present invention may suggest a camera installation position, so if the farm is installed the camera module in an inappropriate position Even in this case, it is preferable to correct it.

1 is a conceptual diagram of a system for performing a method according to the present invention.
2 is a conceptual diagram of facilities such as a vinyl house built in a system for performing the method according to the present invention.
3 is a conceptual diagram of a trap camera module for performing the method according to the present invention.
4 to 9 show an execution screen of a program implemented to perform the method according to the present invention.

Hereinafter, a method according to the present invention will be described with reference to the drawings.

1. Description of a system for carrying out the method according to the invention

This will be described with reference to FIG. 1.

A system for performing a method according to the present invention includes a plurality of camera modules for acquiring images, an image collection unit 400 for collecting images, and a diagnostic server 500 for diagnosing the collected images by comparing them with pest images. And a camera control server 600 for proposing a change in the position of the camera module, and a deep learning server 700 for machine learning. Further, the information obtained in the present invention may be provided to the user terminal 800 and/or the expert terminal 900, and conversely, the information input here may be provided to the diagnostic server 500 or the like.

The diagnosis server 500 performs diagnosis using information stored in the pest database 550.

In the pest database 550, the pest image and the pest information are mapped and stored. The pest information includes the pest name, the drug product name, the pesticide product name including the drug product name, and the pesticide use method of the pesticide product name. Referring to Figure 5, the diagnosis pest name is "ulcer disease", pesticide information for this is shown at the bottom of the figure, it can be seen that the drug product name, pesticide product name, pesticide use method is included.

Meanwhile, a crop image and a crop name may be further mapped and stored in the pest database 550. The crop image will be compared with the image secured by the prediction camera module 200 described below. 4, it can be seen that the image secured by the prediction camera module 200 is shown in the center, and the image is compared with the crop image to confirm the crop name as “paprika”.

The information stored in the pest database 550 may be changed or added by learning by the deep learning server 700 or by information input through the expert terminal 900. This will be described later.

Meanwhile, the plurality of camera modules include a trap camera module 100, a prediction camera module 200, and a stray light insect repellent camera module 300.

The trap camera module 100 secures an image of a worm using handouts such as pheromones (see FIGS. 8 and 9). The manned light insect repellent camera module 300 secures the image after attracting insects to the insect repellent net using light (see FIGS. 6 and 7 ). The prediction camera module 200 secures an image capable of identifying a disease of a crop that is not caused by an worm by checking the crop image (see FIGS. 4 and 5 ).

That is, it is preferable to be able to check the pest in real time by applying various types of camera modules.

As shown in FIG. 2, multiple camera modules can be installed at various locations within the facility. The installation location may be XY coordinated and stored in the camera control server 600. In FIG. 2 shown as an example, the XY coordinates of the facilities are within a range of (0, 0) to (20,000, 8,000) and 4 trap camera modules 100, 2 prediction camera modules 200, and 2 manned lights Installation coordinates of the insect repellent camera module 300 are recorded.

As shown in FIG. 3, the trap camera module 100 is, specifically, a trap body 110 of a hollow type, a trap plate 120 such as a stick provided on the lower surface of the trap body 110, and a trap It includes a trap camera 130 for directing the plate 120, and a transmitting/receiving unit 140 that transmits the image captured by the trap camera 130 to the image collection unit 400.

When a worm attracted to an insect handout (not shown) such as pheromone is trapped by being attached to the trap plate 120, the trap camera 130 is checked in an image taken periodically (see FIGS. 8 and 9).

2. Description of the pest diagnosis and control method according to the present invention

Images are automatically acquired periodically from multiple camera modules. The period can be set or changed in the user terminal 800. For example, the cycle may be 1 day, 3 days, or 1 week, but is not limited thereto.

The image periodically acquired by the camera module is transmitted to the diagnostic server 500 through the image collection unit 400.

Next, the diagnostic server 500 compares the transmitted image with a plurality of pest images stored in the pest database 550, identifies the pest image with the highest accuracy, and verifies the pest information mapped thereto. Now, the diagnostic server 500 transmits the identified pest information to the user terminal 800.

It can be confirmed that there is no transmitted image and no pest image with the highest accuracy. There are no pests. In this case, the diagnostic server 500 transmits pest-free information to the user terminal 800.

That is, in the user terminal, images obtained automatically by a plurality of camera modules are automatically analyzed and only the result value is transmitted. For example, if the cycle is set to 1 day, the user can receive information that there is no pest every day at 1 PM, and if the pest has occurred, the corresponding pest information is received within at least 24 hours. As described above, not only the pest name but also the drug product name, the pesticide product name including the drug product name, and the pesticide use method of the pesticide product name are received together.

In another embodiment of the present invention, to improve accuracy, diagnosis may be made after separately identifying the crop name.

That is, the diagnostic server 500 compares the transmitted image with a plurality of crop images stored in the pest database 550 to identify any one crop image with the highest accuracy, and checks the crop name mapped therein.

Next, the diagnosis server 500 selects a plurality of pest images to which the pest name stored in advance is mapped as corresponding to the previously identified crop name. For example, if the crop name is identified as "paprika", only the pest image identified as occurring in "paprika" is pre-filtered. It is the same that the diagnostic server 500 checks the pest image with the highest accuracy by comparing the selected pest images with the transmitted image, and transmits the pest information mapped here to the user terminal 800.

In another embodiment of the invention, two or more pests may be identified.

That is, the diagnostic server 500 compares the transmitted image with a plurality of pest images stored in the pest database 550, checks two or more pest images with the highest accuracy, and checks two or more pest information mapped to each of them. To the user terminal 800.

At this time, it is possible to identify the common drug item name from two or more pest information, and further transmit it to the user terminal 800. For example, if it is confirmed that worm A and worm B occur simultaneously, information on pesticides capable of controlling worm A and pesticides capable of controlling worm B are transmitted, as well as controlling both worm A and worm B. Information about pesticides that can be done is transmitted separately.

3. Deep learning Description of the method

The present invention has the advantage of increasing accuracy as the method is continuously performed by applying deep learning technology. To this end, a plurality of expert groups are set in advance, and a terminal capable of accessing the system according to the present invention, that is, an expert terminal 900 is provided to them, and information for learning is received therefrom.

The diagnosis server 500 transmits the image transmitted from the user and the pest information identified as related to the expert terminal 900.

When another pest name is input from the expert terminal 900 as correct (that is, when the diagnosis of the diagnosis server 500 is inputted incorrectly), the deep learning server 700 transmits the other pest name confirmed by the expert to the transmitted image. It is mapped and stored in the pest database 550. At this time, it is preferable to store together with the probability of applying the preset expert weight (x). Here, the weight is applied to distinguish it from learning by a user (non-expert), which will be described later.

In addition, if another pest name is input in the other expert terminal 900 for the same pest image, this should increase the probability. Therefore, the accumulated probability is stored.

The deep learning may also be performed by the user terminal 800.

Similarly, when the user terminal 800 inputs another pest name, the deep learning server 700 maps the other pest name to the transmitted image and stores it in the pest database 550, but applies the set user weight (y). Save with probability. Here, the user weight y must be set to be lower than the expert weight x to ensure high accuracy. Likewise, if this process is repeated, the probability is cumulative.

4. Explanation of recommended method for changing camera position

Most preferred is to install a camera module to cover all areas in the facility, but for a variety of reasons such as economic problems, only a few camera modules are actually installed. At this time, the installation location is important. For example, when the camera module is installed so that only a specific area of a facility can be photographed, the time of pest detection starting from another area is delayed and cannot be controlled in a timely manner.

The present invention provides information informing when it is desirable to change the installation position of the camera module after taking an image of a certain period or number of times.

As described above, the camera control server 600 stores the installation coordinates of each of a plurality of camera modules.

After a certain period of time or a number of times of video shooting, the camera control server 600 checks the frequency of checking the pest image for each camera module. That is, it is to check the camera module adjacent to the area where more pests are detected. Now, the camera control server 600 transmits information to the user terminal 800 that recommends moving another camera module toward the installation coordinates of the camera module having a high frequency.

For example, such camera position movement recommendation may be performed on a monthly basis. As illustrated in FIG. 2, in a state in which a total of 8 camera modules are installed, if pests are identified at the most frequent frequency in a specific camera module, it is recommended to change the camera module position a little closer to the installation position of the corresponding camera module. The message is sent to the user terminal 800.

By periodically performing such a task, the user can check the area where the pest occurs more frequently, and acquire the image at an early time, thereby ensuring the timeliness of diagnosis and control.

5. Explanation of new pest image and/or method of entering pest information

Pests are newly discovered and new control technologies for known pests are developed. The present invention provides a technique for updating the database in all these cases. It is preferable that such an update is made through the expert terminal 900 for the accuracy of the data.

First, new pests have been discovered, and control is possible using known control technologies (eg, pesticides already commercialized).

After the expert terminal 900 accesses the pest database 550 and inputs a new pest image, the update is performed by mapping and storing pre-stored pest information in the pest database 550.

Next, new control technologies have been developed for known pests.

After the expert terminal 900 accesses the pest database 550 and selects any one pest image already stored there, the new pest information is further mapped and stored in the pest database 550 to update.

Next, a new pest is discovered, and a new control technology is also developed.

After the expert terminal 900 accesses the pest database 550 and inputs a new pest image, the new pest information is further mapped and stored in the pest database 550 to update.

In another embodiment of the present invention, such information may be transmitted to the user terminal 800 and other expert terminals 900 whenever such an update is made.

6. Implementation example

4 to 9 exemplarily show a screen of a program to which the method according to the present invention is applied.

4 and 5 show an image taken by the prediction camera module 200 crops. On the right, as a diagnostic result, crop and pest names are shown with probability. On the lower side, as pest information other than the pest name, the pesticide product name and the pesticide product name including the drug article name and the pesticide usage method retrieved from the pest database 550 are shown.

6 shows an image in which the man-in-light insect repellent camera module 300 photographs a pest, and FIG. 7 shows the pest name diagnosed based on this. The method according to the present invention is applied, and the pest is identified.

8 shows an image captured by the trap camera module 100. The sticky plate shown in yellow is the trap plate 120. It can be confirmed that the pest name was automatically diagnosed. 9 shows pest information identified based on this. The pest name is shown on the right side, and the lower side shows the drug product name, the pesticide product name including the drug item name, and the pesticide usage method retrieved from the pest database 550.

As described above, the present specification has been described with reference to the embodiments illustrated in the drawings so that those skilled in the art can easily understand and reproduce the present invention, but these are merely exemplary, and those skilled in the art can make various modifications and equivalents from the embodiments of the present invention. It will be understood that embodiments are possible. Therefore, the protection scope of the present invention should be defined by the claims.

100: trap camera module
110: trap body
120: trap plate
130: trap camera
140: transceiver
200: prediction camera module
300: man-in-light insect repellent camera module
400: image collection unit
500: diagnostic server
550: pest database
600: camera control server
700: Deep Learning Server
800: user terminal
900: expert terminal

Claims (15)

(a) transmitting images acquired from a plurality of camera modules to the diagnosis server 500 through the image collection unit 400;
(b) the diagnostic server 500 compares the images transmitted in the step (a) with a plurality of pest images stored in the pest database 550, identifies the pest image with the highest accuracy, and the pest mapped therein Checking information; -Pest information includes pest name
(c) the diagnosis server 500 transmitting the identified pest information to the user terminal 800;
(d) the diagnosis server 500 transmitting the image transmitted in the step (a) and the pest information identified in the step (b) to the expert terminal 900; And
(e) When a pest name other than the pest name included in the pest information is input from the expert terminal 900, the deep learning server 700 maps the other pest name to the image transmitted in the step (a). The step of storing in the pest database 550, and storing with the probability of applying the expert weight (x) is set,
When the above step (e) is repeated in different expert terminals 900 for the same pest image, the probability is increased by accumulating,
Method for diagnosing and controlling pests using the camera module.
According to claim 1,
After step (c),
(c1) If a pest name other than the pest name included in the pest information received by the user terminal 800 is input, the deep learning server 700 may include the other pest name in the image transmitted in step (a). Mapping and storing in the pest database 550, but further comprising the step of storing with the probability of applying the set user weight (y),
When the step (c1) is repeated in different user terminals 800 for the same pest image, the probability is cumulatively increased,
The preset user weight (y) is lower than the preset expert weight (x),
How to diagnose and control pests using camera modules.
According to claim 1,
Step (b) is,
(b1) the diagnostic server 500 compares the images transmitted in the step (a) with a plurality of crop images stored in the pest database 550 to identify any crop image with the highest accuracy, and here Identifying the mapped crop name;
(b2) The diagnostic server 500 selects a plurality of pest images mapped with the pest name stored in advance as corresponding to the crop name identified in step (b1), and transmits the selected pest images and the step (a). Checking the pest image with the highest accuracy by comparing the images; And
(b3) the diagnostic server 500 comprising the step of checking the pest information mapped to the pest image identified in step (b2),
How to diagnose and control pests using camera modules.
According to claim 1,
In step (b), when the diagnostic server 500 confirms that there is no pest image with the highest accuracy and the image transmitted in step (a),
In step (c), the diagnostic server 500 further includes transmitting pest-free information to the user terminal 800,
How to diagnose and control pests using camera modules.
The method of claim 4,
The steps (a) to (e) are repeated every predetermined period, and the preset period can be changed through the user terminal 800,
How to diagnose and control pests using camera modules.
The method of claim 5,
The installation coordinates of the plurality of camera modules are stored in the camera control server 600,
After the above steps (a) to (e) are repeated,
The camera control server 600 checks the frequency at which the pest image is identified for each camera module, and transmits information recommending moving the plurality of camera modules toward the installation coordinates of the camera module having high frequency to the user terminal 800 ) Further comprising the step of transmitting,
How to diagnose and control pests using camera modules.
According to claim 1,
The plurality of camera modules include a trap camera module 100, a prediction camera module 200, and a stray light insect repellent camera module 300,
How to diagnose and control pests using camera modules.
The method of claim 7,
The trap camera module 100,
A hollow trap body 110, a trap plate 120 provided on an upper side of the lower surface of the trap body 110, a trap camera 130 pointing to the trap plate 120, and the trap camera 130 It includes a transmitting and receiving unit 140 for transmitting the captured image to the image collection unit 400,
The image obtained in step (a) includes an image captured by the trap camera 130,
How to diagnose and control pests using camera modules.
According to claim 1,
The pest information includes the pest name, drug product name, agrochemical product name including the drug product name, and a pesticide use method of the agrochemical product name,
How to diagnose and control pests using camera modules.
The method of claim 9,
(f1) the expert terminal 900 further inputting a pest image into the pest database 550; And
(g1) further comprising the step of mapping and storing the pest information previously stored in the pest database 550 by the expert terminal 900 to the further input pest image,
How to diagnose and control pests using camera modules.
The method of claim 9,
(f2) the expert terminal 900 selecting any one pest image stored in the pest database 550; And
(g2) the expert terminal 900 further comprising the step of mapping new pest information to the selected pest image and storing it in the pest database 550,
How to diagnose and control pests using camera modules.
The method of claim 9,
(f3) the expert terminal 900 further inputting a pest image into the pest database 550; And
(g3) the expert terminal 900 further comprising the step of mapping new pest information to the selected pest image and storing it in the pest database 550,
How to diagnose and control pests using camera modules.
The method according to any one of claims 10 to 12,
When a pest image or pest information is further input in the pest database 550, the inputted content is transmitted to the user terminal 800 and another expert terminal 900,
How to diagnose and control pests using camera modules.
The method of claim 9,
Step (b) is,
(b4) the diagnosis server 500 compares the images transmitted in the step (a) with a plurality of pest images stored in the pest database 550, and identifies two or more pest images with the highest accuracy, respectively. Further comprising the step of identifying the mapped two or more pest information,
How to diagnose and control pests using camera modules.
The method of claim 14,
Step (c) is,
(c2) the diagnosis server 500 transmitting the two or more pest information identified in the step (b4) to the user terminal 800; And
(c3) the diagnostic server 500 further comprising the step of identifying the common drug item names in the two or more pest information identified in step (b4), and further transmitting them to the user terminal 800,
How to diagnose and control pests using camera modules.

Images